A robust position controller design for PM synchronous motor using neural network

A robust position controller for permanent magnet synchronous motor is described in this paper. First, linear quadratic method is employed to design the permanent magnet synchronous motor system approximately linear using field orientation theory. Then, the neural network technique with adaptive learning rates is implemented to make the system insensitive to the uncertainties including parameter variations and external disturbance in the whole control process. Finally, the experimental results verify that the dynamic behaviors of the proposed control systems are robust with regards to uncertainties.